International Journal on Future Revolution in Computer Science & Communication Engineering ISSN: 2454-4248

Volume: 3 Issue: 11 396 – 402

_______________________________________________________________________________________________
Collaborative Tagging and Taxonomy by Vector Space Approach

1 2 3
R. Sathya, Mrs. K.V. Sumathi, Mrs. K. K. Kavitha,
M. Phil Scholar, Department of MCA., M. Phil, Assistant Professor, M.C.A., M.Phil., SET., (Ph.D)., Vice
Computer Science, Department of Computer Science, Principal, Head of the Department of
Selvamm Arts and Science College Selvamm Arts and Science College Computer Science, Selvamm Arts and
(Autonomous) Namakkal (Tk) (Dt) – (Autonomous) Science College (Autonomous)
637003. Namakkal (Tk) (Dt) – 637003. Namakkal (Tk) (Dt) – 637003

Abstract:- Collaborative tagging or group tagging is tagging performed by a group of users usually to support in re-finding the items. The
limberness of tagging allows users to classify their collections of items in the ways that they find useful, but the personalized variety of
expressions can present challenges when searching and browsing. When users can liberally choose tags (users create and apply public tags to
online items as different to selecting terms from a proscribed terminology based on the users feedback), the resulting metadata can consist of
homonyms (the same tags used with dissimilar implication) and synonyms (multiple tags for the same concept) which may direct to
inappropriate connections between items and wasteful searches for information about a subject.
Collaborative tagging requires the enforcement of method that enables users to protect their privacy by allowing them to hide certain
user-generated contents without making them useless for the purposes they have been provided in a given online service. This means that
privacy-preserving mechanisms must not harmfully affect the service truthfulness and usefulness.The proposed approach defends the user
privacy to a certain level by reducing the tags that make a user profile let somebody see partiality toward certain categories of interest or
feedback.
__________________________________________________*****_________________________________________________

INTRODUCTION Data mining is a multidisciplinary field, draws

GENERAL BACKGROUND
Data mining also popularly referred to as
Knowledge Discovery from Data (KDD), is the automated
or convenient extraction of patterns representing knowledge
implicitly stored or catchable in large databases, data
warehouses, the Web, other massive information
repositories, or data streams.
work from areas including database technology, machine
learning, statistics, pattern recognition, information retrieval,
neural networks, knowledge-based systems, artificial
intelligence, high-performance computing, and data
visualization. The discoveries of patterns hidden in large
data sets are focusing on issues relating to their feasibility,
usefulness, effectiveness, and scalability.

Data mining and analysis

Problem Identification Business Strategies

Creation of Analytical Data

Environment Develop Tactics and Programs

Application of Data Mining

Tools

Implementation and Tracking Evaluate and Measure

Fig 1.1 Data Mining Basics

396
IJFRCSCE | November 2017, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________
International Journal on Future Revolution in Computer Science & Communication Engineering
Volume: 3 Issue: 11
_______________________________________________________________________________________________
The overall goal of the data mining process is to
extract information from a data set and transform it into an
understandable structure for further use. Aside from the raw
analysis step, it involves database and data management
aspects, data pre-processing, model and inference
considerations, interestingness metrics, complexity
considerations, post-processing of discovered structures,
DIFFERENT LEVELS OF DATA MINING
Artificial neural networks: Non-linear predictive models
that learn through training and resemble biological neural
networks in structure.
Genetic algorithms: Optimization techniques that use
processes such as genetic combination, mutation, and
Predictive
Data Mining
Descriptive
Fig 1.2 Data Mining Level
Decision trees: Tree-shaped structures that represent sets of
decisions. These decisions generate rules for the
classification of a dataset. Specific decision tree methods
include Classification and Regression Trees (CART) and
Chi Square Automatic Interaction Detection (CHAID) .
CART and CHAID are decision tree techniques used for
classification of a dataset. They provide a set of rules that
you can apply to a new (unclassified) dataset to predict
which records will have a given outcome. CART segments a
dataset by creating 2-way splits while CHAID segments
using chi square tests to create multi-way splits. CART
typically requires less data preparation than CHAID.
IJFRCSCE | November 2017, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________
ISSN: 2454-4248
396 – 402
visualization, and online updating. It is useful for computing
science students, application developers, and business
professionals, as well as researchers.
Generally, data mining is the process of analyzing
data from different perspectives and summarizing it into
useful information - information that can be used to increase
revenue, cuts costs, or both.
natural selection in a design based on the concepts of natural
evolution.
Data visualization: The visual interpretation of complex
relationships in multidimensional data. Graphics tools are
used to illustrate data relationships.
Classification
Regression
Time Series Analysis
Prediction
Clustering
Summarization
Association Rule
Sequence Discovery
Nearest neighbor method: A technique that classifies each
record in a dataset based on a combination of the classes of
the k record(s) most similar to it in a historical dataset
(where k 1). Sometimes called the k-nearest neighbor
technique.
Rule induction: The extraction of useful if-then rules from
data based on statistical significance.
PURPOSE OF DATA MINING
The important general difference in the focus and
purpose between Data Mining and the traditional
Exploratory Data Analysis (EDA) is that Data Mining is
more oriented towards applications than the basic nature of
the underlying phenomena.
397
International Journal on Future Revolution in Computer Science & Communication Engineering ISSN: 2454-4248
Volume: 3 Issue: 11 396 – 402
_______________________________________________________________________________________________

Data collection

Data Display

Data reduction

Conclusions: drawing/
verifying

Fig 1.3 Data Mining Usages

Data Mining is often considered to be "a blend of
statistics, AI [artificial intelligence], and data base research
which until very recently was not commonly recognized as a
field of interest for statisticians, and was even considered by
some a dirty word in Statistics.
WEB MINING
Web mining helps to extract useful information
from the web pages. Various we mining techniques are used
to extract knowledge from the web data, web documents and
hyperlinks between the documents. Where the web is
universal information platform space which can be accessed
by companies, universities, businessman etc. Generally,
web hold there are numerous sources of information like
internal sources and external sources.

Fig 1.4 Web Mining

Internal sources are those which include personal
information of any organization and external sources are
those which include information of clients, vendors,
suppliers, intranet and extranet etc. In this research paper,
we can divide us mining into three categories which are
listed as:
a) Web Structure Mining.
b) Web Usage Mining.
c) Web Content Mining.
Web Structure Mining
It consists of web pages as nodes as hyperlinks and
edges connecting related pages. It basically tells the
structural layout of the web. It also used the connectivity
among websites that are called “Hyperlinks”. Hyperlinks
are further divided into two categories which are listed as
below:-
 Internal hyperlinks that lead to pages within the
same web page.
 External hyperlinks that lead to other web pages.
 Document structure is basically a schema language
for XML which helps to describing a valid XML
documents.
Web Usage Mining
It holds the knowledge discovered by users which
are navigating through the websites. files. It is further
divided into two categories which are listed as follows

398
IJFRCSCE | November 2017, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________
International Journal on Future Revolution in Computer Science & Communication Engineering
Volume: 3 Issue: 11
_______________________________________________________________________________________________
 Application Server Data: It holds the business
transactions and also makes their repository in
applications server log.
 Web Server Data: In these logs are made by the
web server. It also includes the field of IP address
means the number of web pages accessed with
access times.
Web Content Mining
It holds the knowledge discovery by going through
the web pages contents like image, videos etc. Intelligent
agents help to solve the problem of indexing in search
engines otherwise it will result in delivery imprecise results
but information overloading. It also helps to select much
more relevant documents.
WEB MINING SECURITY
Web mining has emerged as an important branch of
data mining. This is mainly due to the tremendous amount
of information available from the Web, which attracted
many research communities, and the recent interest of e-
commerce.
OBJECTIVES
 To introduce a user-assisted friend grouping mechanism
that enhances traditional group-based policy
management approaches. Minister to friend grouping
leverages proven clustering techniques to aid users in
grouping their friends more effectively and efficiently.
 To found measurable agreement between clusters and
user-defined relationship groups. In addition, user
perceptions of the improvements should be
encouraging.
 To introduce a new privacy management model that is
an enhancement over conventional group-based policy
management approaches.
 To leverage a user‟s memory and attitude of their
friends to situate policies for other associated friends,
which refer to as Same-As Policy Management.
PROBLEM DEFINITION
Collaborative tagging systems such as Delicious,
last.FM, and Bibsonomy are valuable components of the
Social Web. They allow users, firstly, to organize their own
data with a level of freedom not possible in traditional
taxonomic filing systems whether it is web bookmarks,
music collections, or academic journal references. Secondly,
they provide users a means to openly share this information
so that friends and colleagues can easily communicate with
each other about their latest discoveries. The major issues of
the existing approaches are
 To allow anyone to utilize the collective
knowledge of others for discovering new resources
and perhaps even new friends. These benefits,
however, are only as powerful as the system is
trustworthy. As with any open adaptive system,
IJFRCSCE | November 2017, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________
ISSN: 2454-4248
396 – 402
maintaining the integrity of a tagging system
presents a considerable logistical problem.
 It is loosely classify resources based on end-user‟s
feedback, expressed in the form of free-text labels
(i.e., tags). The novelty of such an approach to
content/resource categorization has been seen, in
recent years, as a challenging research topic.
 The undefined semantics of tags, which are per se
ambiguous and expressed in multiple languages,
makes it difficult to enforce semantic
interoperability and to grant a reasonable level of
accuracy when determining the “meaning” of a tag.
 Tag prediction concerns the possibility of
identifying the most probable tags to be associated
with a non-tagged resource, whereas tag
recommendation is meant to suggest to users the
tags to be used to describe resources they are book
marking. In both cases, existing approaches apply
techniques usually enforced in recommendation
systems.
 Another interesting issue concerns the exploitation
of the “explicit” relationships between users (i.e.,
the actual social network underlying a folksonomy
to address issues like annotation relevance and/or
trustworthiness. Privacy protection in social
tagging services is another issue that has not been
thoroughly investigated.
SCOPE OF RESEARCH WORK
This proposed system‟s methodology could be
implemented in the following real world applications:
 Marketing strategy analysis
 Social communication application of a individual
organization
 Rating process over online
An enhanced collaborative tagging system that
consists of a “traditional” book marking service, such as
Delicious, and two main additional services built on top tree
view. Such services address two main issues. The former
allows end users to specify policies that can be used either to
explicitly denote resources of interests or to enforce
blocking conditions on the browsed data. The latter features
a specific PET, namely, tag suppression, to preserve the
privacy of registered users by hiding the specific
characteristics of their profiles. Such architecture is a
specific implementation of the multilayer framework
presented. with the relevant difference that in [5] the privacy
layer is missing. Lastly, we would also like to emphasize
that our approach is not limited to the specific book marking
application here contemplated, i.e., Delicious. As a matter of
fact, it could be built on top of any collaborative tagging
system.
Nevertheless, if tags were not sensible information
per se, they could easily be exploited to infer users‟ personal
399
International Journal on Future Revolution in Computer Science & Communication Engineering
Volume: 3 Issue: 11
_______________________________________________________________________________________________
information, such as personal interests, preferences, and
opinions. This is even easier when it is possible to
statistically analyze huge collections of tags as those made
publicly available by social bookmarking services, thus
obtaining accurate tag-based user profiles. In this field,
privacy-preserving techniques should guarantee, at the same
time privacy protection and the correctness of the results
obtained by analyzing the data set.
LITERATURE REVIEW
Wu, L. Zhang, and Y. Yu et al[2] proposed explore
a complement approach that focuses on the"social
annotations of the web" which are annotations manually
made by normal web users without a pre-defined formal
ontology. Compared to the formal annotations, although
social annotations are coarse-grained, informal and vague,
they are also more accessible to more people and better
reflect the web resources' meaning from the users' point of
views during their actual usage of the web resources. But
this system focuses only on bookmarking web based system.
Also this system not considers the tagging approach.
B. Markines, C. Cattuto, F. Menczer, D. Benz, A.
Hotho,and S. Gerd et al[3] proposed Here they build an
evaluation framework to compare various general
folksonomy-based similarity measures, which are derived
from several established information theoretic, statistical,
and practical measures. Their framework deals generally and
symmetrically with users,tags, and resources. For evaluation
purposes we focus onsimilarity between tags and between
resources and consider different methods to aggregate
annotations across users.This approach shows that they can
define relations of users from tags. This is important from
privacy preserving point of view.
C. Marlow, M. Naaman et al[4] provide a short
description of the academic related work to date. They offer
a model of tagging systems, specifically in the context of
web-based systems, to help us illustrate the possible benefits
of these tools. Since many such systems already exist, they
provide a taxonomy of tagging systems to help inform their
analysis and design, and thus enable researchers to frame
and compare evidence for the sustainability of such systems.
They also provide a simple taxonomy of incentives and
contribution models to inform potential evaluative
frameworks. They present a preliminary study of the photo-
sharing and tagging system Flicker to demonstrate our
model and explore some of the issues in one sample system.
Hence this paper is just giving us the basic idea about how
tag functionality works in web based systems. This is
important to clear basic ideas about tagging.
METHODOLOGY
INTRODUCTIONTAG
A tag is a user-contributed metadata, providing a mean
of information or content item, created freely by users with
personally salient keywords or labels, known as tags. The
IJFRCSCE | November 2017, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________
ISSN: 2454-4248
396 – 402
process of labeling is called tagging. Users can re-find the
information later by means of those tags that they have
created. Also, by tagging users can store resources for their
future retrieval.
COLLABORATIVE TAGGING SYSTEM
Collaborative tagging is a classification by the users and
for the users. It is a social, decentralized and complex
network where many annotations, generally provided by
interrelated groups of individuals, are organized so to link
resources and tags. Each resource item can be associated
with many different tags, rather than with a single branch of
a hierarchy.
With tags chosen freely from common language and
associated with web resources that are interesting for users
(such as photographs, videos, web links and documents),
collaborative tagging offers a sense of community in
managing resources and results in a process of knowledge
construction. Users can share their resources with others,
discover resources through the collaborative network, and
contact people with similar interests. The benefit of
collaborative tagging systems comes from the many views
of the mass, rather than from a dominant opinion supplied
by a few.
In fact, the form of tagging tends to stabilize over time
because people usually choose to use the tags in three ways:
 Imitation, users are easily affected by the tags
that were previously applied by others to the
same page;
 Habit, users re-use tags that they have already
used on other pages respect to their
background and culture;
 Recommendation, users choose tags that are
suggested by a given interface.
COLLABORATIVE METHODS
Several recommendation systems use a hybrid approach
by combining collaborative and content-based methods,
which helps to avoid certain limitations of content-based and
collaborative systems. Different ways to combine
collaborative and content-based methods into a hybrid
recommender system can be classified as follows:
 Implementing collaborative and content-based
methods separately and combining their
predictions,
 Incorporating some content-based characteristics
into a collaborative approach,
 Incorporating some collaborative characteristics
into a content-based approach, and
 Constructing a general unifying model that
incorporates both content-based and collaborative
characteristics.
TAG SUPPRESSION
400
International Journal on Future Revolution in Computer Science & Communication Engineering ISSN: 2454-4248
Volume: 3 Issue: 11 396 – 402
_______________________________________________________________________________________________
In our scenario of collaborative tagging, users tag  It is useful for developing software components
resources on the web, for example, music, pictures, videos which are suitable for deployment in the distributed
or bookmarks, according to their personal preferences. Users environments. This supports internationalization.
therefore contribute to describe and classify those resources, CHARACTERISTICS OF C#
but this is inevitably at the expense of revealing their profile.  Garbage Collection: the memory management
To avoid being accurately profiled by tagging systems or in feature leads all managed objects. Garbage
general by any attacker able to collect such information, collection is a feature .NET. The C# uses it during
users may adopt a privacy-enhancing technology based on the runtime.
data perturbation. The data-perturbative technology  Indexes: C# has indexes which help to access
considered in this work is tag suppression, a technique that value in a class with an array like syntax programs.
allows a user to refrain from tagging certain resources in  Exception Handling: .NET standardizes the
such a manner that the profile resulting from this exception handling across languages. C# offers the
perturbation does not capture their interests so precisely. conditional keyword to control the flow and make
PROPOSED METHOD PROCEDURE the code more readable.
The proposed methodology of the thesis is  Versioning: C# programming supports this
implemented with the following modules and procedures to versioning. The .NET solves the versioning
experiment the proposed methodology. problem and enables the software developer to
 User Registration specify version dependencies between the different
 Post Content pieces of software.
 View Posts  Extensive Inter-operability: All enterprise
 General Suppression Word software application can be managed easily by type
 Strict Suppression Word safe environment. This extensive inter-operability
 My Tag Cloud makes C# which is the obvious choice for the
 Add Child User software developers.
 View Child Users EXPERIMENTAL RESULTS
 Policy/ Resource Recommendation INPUT FORM DESIGN USER LOGIN PAGE
 Policy/ Parental Control
 Change Password
EXPERIMENTAL RESULTS AND DISCUSSION
IMPLEMENTATION SOFTWARE
The empirical system is designed and implemented
by using the Microsoft visual studio .net as a front end tool.
And the coding language used is C#.net. Microsoft SQL
Server used as a back end tool. Visual studio is an integrated
development environment which is used in this thesis for
designing the thesis experiments.
FEATURES OF C#.NET
C# is a Microsoft„s new language designed for its
USER OPTION
new platform “.NET”. It is fully object oriented language
like java and is the first component-oriented language.
Because it contains integral supports for writing the software
components. C# is designed for building robust, reliable and
durable components to handle real world application. The
C# language specification stated the objectives and features
of C#:
 It is simple, modern, general purpose and object
oriented programming language.
 This provides a support for the software analysis
principles such as strong type checking, array
bounds checking, detection of attempts to use
uninitialized variables and automatic garbage
collection.

401
IJFRCSCE | November 2017, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________
International Journal on Future Revolution in Computer Science & Communication Engineering ISSN: 2454-4248
Volume: 3 Issue: 11 396 – 402
_______________________________________________________________________________________________
CONCLUSION Collaborative Computing: Networking, Applications and
Collaborative tagging is currently an enormously Work sharing, pp. 126-144, 2008.
popular online service. Although nowadays it is basically [6] Frı´as-Martinez.E, Cebria´n.M , and Jaimes.A , “A Study
on the Granularity of User Modeling for Tag Prediction,”
used to support resource search and browsing, its achievable
Proc. IEEE/WIC/ACM Int‟l Conf. Web Intelligence
is still to be demoralized. One of these potential applications
Intelligent Agent Technology (WIIAT), pp. 828-831,
is the provision of web access functionalities such as content 2008.
filtering and innovation. For this to become a reality,
however, it would be necessary to extend the architecture of
current collaborative tagging services so as to include a
policy layer that supports the enforcement of user
inclinations.
Collaborative tagging has been gaining popularity,
it have been become more obvious the need for privacy
safeguard; not only because tags are susceptible information
but also because of the risk of cross referencing. In addition
to the existing system approaches, the proposed system
takes care of multi language tagging.

FUTURE ENHANCEMENTS
A privacy preserving collaborative tagging if
functional to content with various languages, and then it
becomes more efficient to beneficial to end users. Future
work includes the development of a full prototype for the
experimented system and it‟s testing and use in further
scenarios.
The proposed methodology can be enhanced and
implemented with the various applications like
 Official information sites,
 Employees skills registry portals,
 Other official or personal related
confidential information management
system through online.
 In future the algorithm will be modified or
redefined to improve the efficiency of the
methodology.
REFERENCES
[1] Adomavicius.G and Tuzhilin.A, “Toward the Next
Generation of Recommender Systems: A Survey of the
State-of-the-Art and Possible Extensions,” IEEE Trans.
Knowledge Data Eng., vol. 17,no. 6, pp. 734-749, June
2005.
[2] Barnes.S.B, “A Privacy Paradox: Social Networking in
the United States,” First Monday, vol. 11, no. 9, Sept.
2006.
[3] Bischoff .K ,Firan . C . S, Nejdl . W, and Paiu . R , “Can
All Tags Be Used for Search?” Proc. 17th ACM Conf.
Information and Knowledge Management (CIKM), pp.
193-202, 2008.
[4] Bundschus.M, Yu.S, Tresp.V, Rettinger.A , Dejori.M ,
and Kriegel.H.P, “Hierarchical Bayesian Models for
Collaborative Tagging Systems,” Proc. IEEE Int‟l Conf.
Data Mining (ICDM), pp. 728-733,2009.
[5] Carminati.B, Ferrari.E , and Perego.A, “Combining
Social Networks and Semantic Web Technologies for
Personalizing Web Access,” Proc. Fourth Int‟l Conf.
402
IJFRCSCE | November 2017, Available @ http://www.ijfrcsce.org
_______________________________________________________________________________________